Patriarchy at the ballot box: Public perceptions of female leadership and the limits of democratic consolidation in pre-election Indonesia (2024)

Abstract

Despite three decades of affirmative-action legislation and a steady rise in Indonesia's Gender Empowerment Index, the highest office in the world's third-largest democracy has been held by a woman only once. This article asks why, drawing on an original nationwide survey conducted on the eve of the 2024 presidential election. Using a multistage random sampling design across all 34 provinces, the study generated a representative sample of 1240 voting-age citizens (margin of error ± 5.7 per cent at the 95 per cent confidence level; DEFF = 3.89), allowing for the first systematically comparable measurement of public attitudes toward a hypothetical female head of state in pre-2024 Indonesia. The findings reveal a divided but tilted electorate: 51.9 per cent reject the principle of a female presidential candidacy, while 62.4 per cent say they would not vote for one in practice. This ten-point gap—what we term permissive bias —captures an attitudinal architecture in which principled endorsement systematically exceeds stated behavioural willingness. A McNemar's test confirms that this gap is statistically significant (χ 2 (1) = 97.656, p < 0.001), with a directional asymmetry of 143 discordant cases in the permissive direction versus only 17 in the reverse direction. Disaggregation by sex, age, and monthly household expenditure, supported by chi-square tests and binary logistic regression, reveals that sex is the sole demographic variable significantly associated with both outcome measures (principled agreement: χ 2 (1) = 35.094, p < 0.001; voting intent: OR = 1.831, p < 0.001). Men oppose female candidates at substantially higher rates than women do, and this pattern holds across all age cohorts. Age, expenditure, and party affiliation show no statistically significant associations; education is modestly associated with principled agreement only (χ 2 (7) = 16.766, p = 0.019). We emphasise that demographic variables jointly contribute only a modest amount to model fit (McFadden Pseudo-R 2 = 0.019); sex is best understood as a dominant demographic predictor rather than a dominant substantive determinant. We interpret these patterns primarily through Shestopal's multi-layer approach to political perception, with feminist scholarship on gendered leadership (Connell, 2005; Eagly & Carli, 2007), intersectionality (Crenshaw, 1991; Hancock, 2007; Krook & Norris, 2014), and feminist institutionalism (Kantola & Lombardo, 2017) providing supporting frames. Because our measures are stated attitudes rather than observed voting behaviour, all inferences about ballot-box outcomes are attitudinal in scope. The article contributes the first nationally representative attitudinal baseline for Indonesia. It offers permissive bias as a conceptual tool for comparative research on the disjuncture between egalitarian norms and exclusionary electoral attitudes worldwide.

The authors' abstract, as published at the source. Social Sciences & Humanities Open, 2026 · DOI ↗